[sdiy] 0.100 pin header reliability.

[David G. Dixon]

> Ah, ok, the cheap $10 ones :)
> 
> Crimping is only guaranteed to work as stated by the manufacturer when
> (a) you use the correct crimping tool, and (b) that tool is within
> calibration (i.e., it needs regular checking and recalibration if
> required).  If you spend $300 or more on a crimp tool then you're in
> the right ballpark.

My crimping pliers have two crimping locations: A and B.  One is a bit
smaller, and that's the one I use.  The two sides of the pliers come
together very tightly -- these pliers are probably at least 30 years old and
very well made.  I haven't seen any this nice in the stores, but I'm sure
they'd go for between $50 and $100.

The crimp pins I use have two sets of crimpable "wings" -- smaller ones in
the middle for crimping onto bare wire, and larger ones on the end for
crimping onto the insulation.

I strip about 1/8" of insulation off the end of the wire, put it in the pin
and crimp the small wings hard with the crimping pliers.  (I take pains to
ensure that all the little wire strands get crimped in, although very
occasionally a stray strand gets away.)  Then I crimp the large wings over
the insulation.  The pin is on that wire so damned hard that you could never
pull it off.

Once the PCB is attached to the panel and all the connectors are plugged in,
there is no pulling stress on the wires, and very little if any movement is
possible since I bundle them together with plastic ties.

So, what is going to fail here?  What does the $300 dollar machine do that
my crimping pliers don't?